|
Title:
|
APPLICATION OF MARKOV CHAIN TO PREDICT FAULTS FROM REAL TIME ALARM DATA |
|
Author(s):
|
Ahmad Kazmi |
|
ISBN:
|
978-972-8939-19-9 |
|
Editors:
|
Hans Weghorn, Jörg Roth and Pedro Isaías |
|
Year:
|
2010 |
|
Edition:
|
Single |
|
Keywords:
|
Alarm prediction, Faults prediction, Telecommunication system, Markov Chain |
|
Type:
|
Short Paper |
|
First Page:
|
197 |
|
Last Page:
|
201 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Current telecommunication systems are complex heterogeneous networks with parts from various vendors. Obviously such complex systems will face a number of faults that may deny services to the end users, hence resulting in revenue losses to the telecommunication companies. Various techniques are proposed that predict fault based on historical alarm data. One such technique is from the Telecom Alarm Sequence Analyzer (TASA) project that has proposed to classify the sequence of alarms into 3 categories. We have used TASA techniques to categorize alarms and then Markov Chain technique for classification of future sequence of alarms according to one of TASA categories. Once we know the category of a sequence of alarms we can predict the alarm (fault) itself. We have applied our proposed method on real time alarm data of a telecommunication company. Furthermore, we present the alarm prediction results to verify that our approach has merit. |
|
|
|
|
|
|
